2026 May 08 #
2026 May 07 #
- You Should Never Be The Most Sycophantic Participant In A Conversation With A Chatbot - "Imagine a film director telling an actor to play a scene with greater emotive intensity, and then afterward being like 'Jeez, I'm so sorry to have upset you.' Imagine a costume designer dressing a performer up like Albert Einstein and thinking that would make them capable of explaining general relativity. Imagine a gamer turning up the difficulty setting on FIFA and thinking they'd made their Playstation better at soccer.
Andreessen is creating—typing out and entering, but not into the chatbot—his own delusion. In trying to tell the chatbot not to hallucinate, he is scripting his own psychotic break. He is doing it because he is a huge dumbass. Don't expect Claude to tell him so."
- Life During Class Wartime - "War is bad. Don’t start one. But we’re already in a class war and we’re losing. Where by 'we' I mean most people; the winning side comprises, roughly, the richest 0.1% of the population, who are morphing into a hereditary aristocracy."
- This Canadian millionaire wants to pay higher taxes
2026 May 05 #
- What Sort of AI Bubble Are We In? - “One of the major reasons why existing data centers cannot keep up with the computational demand is that the AI industry is not the least bit interested in pursuing efficiencies that would make their products less computationally demanding. They’ve constructed a sort of financial flywheel, where they subsidize computation-heavy behavior from users, then report back to investors about the soaring product demand and use it to justify the next investment round. If they built more efficient AI models, then it would paradoxically become harder to attract the next trillion dollars of investor cash.”
2026 Apr 30 #
- The Social Edge of Intelligence - ‘AI doesn’t really “think.” Rather, it remembers how we thought together. And we’re about to stop giving it anything worth remembering.’
2026 Apr 23 #
2026 Apr 21 #
- AI has limits, even if many AI people can't see them - “He describes some famous results from the research of psychologist Paul Meehl on medical and other decisions, which suggested that 'statistical prediction provided more accurate judgments about the future than clinical judgments' under certain conditions. But the conclusion that Ben comes to is not that this means that statistical prediction is generally better than expert judgment. Instead, it is better when there are clearly defined outcomes, good data, and clear reference cases that can be used for comparison. There are many situations in which this is not true, and cannot readily be made true.”
- everything is a nail, or at least it ought to be - “As he puts it (following Paul Meehl), algorithmic decision making is always going to have the evidence on its side. Because once you have put the problem in terms of the kinds of things which can be measured and defined a specific success metric - once there is any standard of evidence with which to judge the results - then 'optimisation' means what it says. Anything you do differently from the output of an optimiser is … suboptimal.
But this often means that all the work is done in deciding what to measure and what the optimand should be, what counts as evidence and what as a test. Not only is that process a great way to put your thumb on the scale without leaving fingerprints, a lot of the time things get measured because they are convenient to measure, rather than any particularly principled reason. As I’ve constantly said in econometric context, the easiest way to find a valid instrument for an unobservable quantity is simply to lower your standards.”
- LLMe - An interesting observation here is that LLM code ingestion doesn’t appear to distinguish good from bad examples, and tends towards examples that work at all, but both good and bad have a lot to teach if we keep clear which one we’re looking at.
2026 Apr 16 #
- The peril of laziness lost - “The problem is that LLMs inherently lack the virtue of laziness. Work costs nothing to an LLM. LLMs do not feel a need to optimize for their own (or anyone’s) future time, and will happily dump more and more onto a layercake of garbage. Left unchecked, LLMs will make systems larger, not better — appealing to perverse vanity metrics, perhaps, but at the cost of everything that matters. As such, LLMs highlight how essential our human laziness is: our finite time forces us to develop crisp abstractions in part because we don’t want to waste our (human!) time on the consequences of clunky ones. The best engineering is always borne of constraints, and the constraint of our time places limits on the cognitive load of the system that we’re willing to accept. This is what drives us to make the system simpler, despite its essential complexity.”
- Zen fascists will control you... - "Jello Biafra was writing a joke about a California politician.
He was also writing a warning about a kind of person.
That kind of person, today, is doing very well for themselves indeed."
2026 Apr 09 #
2026 Apr 08 #
2026 Apr 04 #
- The machines are fine. I'm worried about us. - "He said that 'LLMs will take away what's so great about science.' At the time, I thought he was just talking about his own competitive edge, his fluency as a native English speaker, his ability to write fast and publish often. And he was. But I've come to think the phrase itself was more right than he knew, even if his reasons for saying it were mostly self-interested. What's great about science is its people. The slow, stubborn, sometimes painful process by which a confused student becomes an independent thinker. If we use these tools to bypass that process in favor of faster output, we don't just risk taking away what's great about science. We take away the only part of it that wasn't replaceable in the first place."
2026 Apr 03 #
2026 Mar 26 #
2026 Mar 25 #
2026 Mar 24 #
2026 Mar 23 #
- Slop-Machine Future • Buttondown
- Sucker: My year as a degenerate gambler
- "Gooning Towards The Führer" as policy coordination - “The pathologies of Seeing Like a State - of failing to observe or understand problems that cannot be broken down into simple metrics and regularized categories are very well documented. The state can’t easily coordinate its activities to deal with problems that cannot be expressed in terms that it understands. The pathologies of Seeing Like an Idiot are worse understood, because we haven’t had to think so hard about them. They don’t just involve ineptitude and blunders, but a nearly complete incapacity to see or talk about the much wider set of problems that can’t be expressed via goonability. By and large, the abstractions of goonability carry sparser and less useful information than the abstractions of bureaucracy.”
2026 Mar 07 #
- Stone Tools - "Exploring retro productivity software from the 8/16-bit era. No games, just work."
2026 Feb 08 #
2026 Feb 04 #
- LLMs at Oxide - This sort of you-need-to-hold-it-right advice reminds me of learning to hit Cmd-S every 30 seconds or so without conscious thought, as a key skill of using a 1990s Mac.
2026 Jan 13 #
2025 Dec 11 #
- What are formal systems? - I'm very charmed by the idea that just the right mix of meaning and meaninglessness can generate power in the world.
2025 Dec 10 #
2025 Dec 01 #
- Is there a general skill of “management”? - "[T]he kernel of my argument for the existence of a general skill of management is that it is pretty obvious that there is a general deficit or 'negative skill' of _mismanagement_, which equally obviously appears to work in roughly the same way in a variety of fields, and that therefore an opening stab at a definition of the general skill of management would be that it’s the absence of this deficit."
2025 Nov 26 #
- Large Language Models and Emergence: A Complex Systems Perspective - “Minimally, emergence describes the reorganization of a system that can support a new, often smaller description, that screens off microscopic details not essential to predicting the future of a system. Emergence matters because it leads to an enormous cost saving in how systems are described, predicted, and controlled. We do not need to use quantum mechanics to build a bridge because the classical world emerges from the quantum—a fact exploited by engineers.”
- My Life Is a Lie - “Altruism is a function of surplus. It is easy to be charitable when you have excess capacity. It is impossible to be charitable when you are fighting for the last bruised banana.”
- The Trillion-Dollar Cost of IT’s Willful Ignorance - "It may be a forlorn request, but surely it is time the IT community stops repeatedly making the same ridiculous mistakes it has made since at least 1968, when the term “software crisis” was coined. Make new ones, damn it."
2025 Nov 18 #
- A Month of Chat-Oriented Programming - "[...] I got Claude to write well over 99% of the code produced during the month. I found the experience infuriating, unpleasant, and stressful before even worrying about its energy impact. Ideally, I would prefer not to do it again for at least a year or two. The only problem with that is that it “worked”. [...] Against my expectation going in, I have changed my mind. I now believe chat-oriented programming (“CHOP”) can work today, if your tolerance for pain is high enough."
2025 Nov 13 #
2025 Oct 21 #
- On AI and the golem - “We are teetering on the edge of the collapse of this particular bubble, and there’s no doubt that ordinary people who didn’t waste precious resources generating sub-par horny images are going to pay the price. Unfortunately, there is very little I can do for us in that regard. What I can do is highlight the fact that the people of the past already answered the philosophical questions surrounding this particular bad idea for us. As we are constantly told that the arts and humanities have no value, and that we can have machines do that work for us, I think this is an important reminder. To be human is to do the work. For better or worse.”
2025 Oct 17 #
- Choosing friction - That feeling when you're pretty sure you've found one of your people.
- Right Turn of Death - What the what the right-turn-on-a-red-light is _younger than I am_ and was a largely-ineffective idea to save gas during the oil shocks of the 1970s.
2025 Oct 15 #
2025 Oct 09 #
- Who Goes Nazi? - From 1941: “It is an interesting and somewhat macabre parlor game to play at a large gathering of one’s acquaintances: to speculate who in a showdown would go Nazi.”
- Dewaffling the tech industry
2025 Oct 08 #
2025 Jun 25 #
2025 Jun 08 #
2025 Jun 07 #
2025 May 15 #
2025 May 04 #
- How to Structure a Clojure Web App 101 - “What has been a challenge is explaining what exactly it is that these libraries do. Doing that - really doing that - requires a mountain of shared context that folks simply do not have.”
2025 Apr 04 #
- On the foolishness of 'natural language programming' - “It was a significant improvement that now many a silly mistake did result in an error message instead of in an erroneous answer. (And even this improvement wasn't universally appreciated: some people found error messages they couldn't ignore more annoying than wrong results, and, when judging the relative merits of programming languages, some still seem to equate "the ease of programming" with the ease of making undetected mistakes.)”
2025 Feb 21 #
- HUMAN_FALLBACK - A wonderfully-written memoir of time spent being behind the curtain of a Potemkin AI.
- Why Clojure? - An overview of what makes it a pleasure to work in.
2025 Feb 11 #
- Situated Software - Re-reading Clay Shirky articles I first read before the Facebook era just hurts. I hear a Ron Howard Arrested Development Narrator voice in the back of my head saying, "But unfortunately, Facebook."
- An app can be a home-cooked meal - If “learn to code” were meant the same way as “learn to cook.”
2025 Feb 06 #
- You Can’t Post Your Way Out of Fascism - “You can discourse and quote-dunk and fact-check until you’re blue in the face, but at a certain point, you have to stop and decide what truth you believe in. The internet has conditioned us to constantly seek new information, as if becoming a sponge of bad news will eventually yield the final piece of a puzzle. But there is also such a thing as having _enough_ information.”
2025 Feb 02 #
2025 Jan 31 #
- CodeCharta - Codebases visualized as architectural models seems very much like my jam.
- A Rant about "Technology" - “Technology is the active human interface with the material world. But the word is consistently misused to mean only the enormously complex and specialised technologies of the past few decades, supported by massive exploitation both of natural and human resources. This is not an acceptable use of the word.”
2025 Jan 30 #
2025 Jan 27 #
2025 Jan 26 #
- JAR on 'Object-Oriented' - "Because OO is a moving target, OO zealots will choose some subset of this menu by whim and then use it to try to convince you that you are a loser."
- In defense of blub studies - "So I always feel a little bit embarrassed and boring when I instead suggest going really deep on what you already know: your main programming language, web framework, object-relational mapper, UI library, version control system, database, Unix tools, etc. It’s not shiny or esoteric, but for me, building a detailed mental model of those (and how they compare to alternatives) might be the learning that’s contributed most to my effectiveness as an engineer."
2024 Dec 28 #
2024 Dec 27 #
- Ghostty - Ghostty came out yesterday, if you're into terminal emulators.
- filipesilva/datomic-pro-sqlite - Datomic is so abstract on the storage end that it took me a long time to figure out how to ... store stuff. Thus, I am really liking this docker image for quickly setting up a transactor that uses sqlite. (For low to medium write volumes this seems fine, since architecturally nothing is competing with the transactor for writes anyway.